Politics & Government

How could the polls get Tuesday’s election so wrong?

Almost all the forecasts and polls, and virtually all the pundits on the internet agreed Tuesday morning that the United States was headed for a Hillary Clinton administration.

For months, Democrat Clinton consistently had led in Election Day forecasts. But the end result early Wednesday morning was very different.

Republican Donald Trump was elected president with at least 279 electoral votes, a result almost no one saw coming.

How could almost all the experts be wrong?

Trump won in Pennsylvania and Wisconsin, and, as of Wednesday, led in Michigan, three Rust Belt states that weren’t supposed to be in play for Republicans.


As late as Tuesday morning, the number-crunchers at FiveThirtyEight were touting Clinton as a 71 percent favorite to win the presidency. “There’s A Wide Range Of Outcomes, And Most Of Them Come Up Clinton” was the headline on statistician Nate Silver’s last election article.

As the headline indicates, Silver — unlike some other election predictors — emphasized throughout the campaign that the polls were close and the large number of undecided voters gave Trump about a one-in-three chance of victory.

Silver’s site had predicted correctly every state in the 2012 presidential election, and its final electoral map predicted Clinton would finish Tuesday with 323 electoral votes.

With some states still not decided Wednesday, Clinton only had managed to get 228 electoral votes.

After the results became clear, Silver offered his take on Trump’s election.

“In an extremely narrow sense, I’m not that surprised by the outcome, since polling – to a greater extent than the conventional wisdom acknowledged – had shown a fairly competitive race with critical weaknesses for Clinton in the Electoral College,” he said. “But in a broader sense? It’s the most shocking political development of my lifetime.”

Larry Sabato, director of the University of Virginia’s Center for Politics, was more explicit in his post-election blog post, entitled “Mea Culpa, Mea Culpa, Mea Maxima Culpa.”

It’s first line? “What can I say? We blew it.”

Sabato’s forecast had projected a Clinton win with 322 electoral votes, a forecast he admits missed Trump’s winning appeal in four Democratic-leaning states that the Republican carried and two more that were too close to call even Wednesday.

“We thought the signs pointed to Hillary Clinton winning the White House. We thought that even if she lost Florida, North Carolina and Ohio, her Midwestern ‘firewall’ of states that not only had voted for Barack Obama twice, but hadn’t voted for a Republican since the 1980s, would hold for her. It didn’t – Trump blew a hole in what we dubbed ‘Fortress Obama.’ ”

Real Clear Politics was closer than the others.

Its final national polling average gave Clinton a 3.2-percentage-point lead nationally, a full 3 points higher than the 0.2 percent lead in the popular vote that the Democrat had Wednesday. (Yes, Clinton apparently won the popular vote even though she lost the election.)

Real Clear projected a slim 272 electoral vote win for Clinton, a lead that depended on the Democrat holding the “firewall” of Michigan, New Hampshire, Pennsylvania and Wisconsin.

She didn’t.

Others were more optimistic.

The Upshot blog at the New York Times gave Clinton an 85 percent chance of victory on Election Day.

But that was modest compared to the Huffington Post’s Pollster, which gave Clinton a 98 percent chance of victory and President-elect Trump a less than 2 percent chance.